The impact of IoT-ML in the healthcare sector is very significant and it has helped us to change our view at the traditional treatment methods. In IoT-ML-based healthcare applications, the sensing layer is responsible for collecting information from humans and transferring it to the storage layer through communication technology. ML is implemented to make intelligent decisions for healthcare applications. This survey shows all the fields starting from the IoT sensor devices to the deployment of ML in the healthcare sector. We have conducted a comprehensive survey of the existing literature covering IoT and ML strategies from a healthcare perspective. We also provide insights into the different types of network storage and computing strategies used for other health-based applications. We believe that the presented work is innovative as no other survey is furnished in such manner. From this survey, researchers can get an overview of IoT-ML and cloud-based healthcare applications under the single system. We have proposed a unique taxonomy from an IoT-ML-based healthcare perspective where we have highlighted key steps in developing healthcare systems. We have culminated the most striking technologies in IoT, communications, network storage and computing, and ML for healthcare systems. Another contribution of our survey is that we have collected and discussed surveys and scientific literature based on the proposed taxonomy and their sub-taxonomy throughout this paper. Besides that we have reviewed several types of popularly used sensors, development boards in healthcare with various examples. We also show the mapping of communication technology with the protocols used by IoT sensors. In the ML section, we have shown an ML pipeline centering on healthcare application and discussed every step of it. Finally, we have identified a number of research challenges including exploration of Deep Learning based models, proper data acquisition and handling of data, privacy and ethics, security issues in WBAN, etc. These research challenges will provide the researchers the necessary future research directions while developing IoT-ML-based healthcare applications.

IoT-ML 在医疗保健领域的影响非常显着，它帮助我们改变了对传统治疗方法的看法。在基于 IoT-ML 的医疗保健应用中，传感层负责从人类收集信息并通过通信技术将其传输到存储层。ML 用于为医疗保健应用做出智能决策。该调查显示了从 IoT 传感器设备到医疗保健领域 ML 部署的所有领域。我们从医疗保健的角度对涵盖物联网和机器学习策略的现有文献进行了全面调查。我们还提供了有关用于其他基于健康的应用程序的不同类型的网络存储和计算策略的见解。我们相信所呈现的工作具有创新性，因为没有其他调查以这种方式提供。通过这项调查，研究人员可以大致了解单一系统下的 IoT-ML 和基于云的医疗保健应用程序。我们从基于 IoT-ML 的医疗保健角度提出了一种独特的分类法，其中强调了开发医疗保健系统的关键步骤。我们在物联网、通信、网络存储和计算以及医疗保健系统的机器学习方面达到了顶峰。我们调查的另一个贡献是我们在整篇论文中收集和讨论了基于提议的分类法及其子分类法的调查和科学文献。除此之外，我们还通过各种示例回顾了多种类型的常用传感器、医疗保健开发板。我们还展示了通信技术与物联网传感器使用的协议的映射。在 ML 部分，我们展示了一个以医疗保健应用为中心的 ML 管道，并讨论了它的每一步。最后，我们确定了一些研究挑战，包括探索基于深度学习的模型、正确的数据采集和数据处理、隐私和伦理、WBAN 中的安全问题等。这些研究挑战将为研究人员提供必要的未来研究方向，同时开发基于 IoT-ML 的医疗保健应用程序。